Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 23323
    Prev
    Next
    INDEX
    Explanations

    expressions of frustration or strong emotion, particularly featuring the word "damn."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    352
    +0.12
    0.6%
    526
    +0.12
    0.5%
    604
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    604
    +0.12
    0.01
    562
    +0.12
    0.01
    352
    +0.11
    0.01
    Negative Logits
    NetMessage
    -0.86
    KY
    -0.79
     Cosponsors
    -0.78
    anwhile
    -0.74
    Effect
    -0.73
    Edit
    -0.70
    IDS
    -0.67
    OHN
    -0.66
    idon
    -0.65
    HCR
    -0.65
    POSITIVE LOGITS
     damned
    0.96
     damn
    0.96
     darn
    0.94
     thing
    0.90
    ation
    0.85
    ibly
    0.83
    near
    0.82
    warm
    0.74
     dirty
    0.73
     goddamn
    0.70
    Activations Density 0.006%

    No Known Activations